H U M A N I M M U N E R E S P O N S E S TO H A P T E N - C O N J U G A T E D CELLS I. P r i m a r y a n d S e c o n d a r y P r o l i f e r a t i v e R e s p o n s e s in Vitro* BY MICHAEL F. SELDIN AND ROBERT R. RICH$ (From the Departments of Microbiology and Immunology and of Medicine, Baylor College of Medicine, and the Methodist and Veterans Administration Hospitals, Houston, Texas 77030)

The association between the human major histocompatibility complex (MHC) 1 and a variety of diseases has been well established in the last few years (1, 2). In contrast with the mouse, however, in which the availability o f H - 2 recombinant strains has permitted rapid progress, relatively little is known about the functional role of MHC genes in human immune responses. In the mouse, genes within specific regions of the H - 2 complex determine susceptibility to certain diseases (3), control immune responses to numerous antigens (4), and determine proliferative and cytotoxic immune responses of T cells (5-8). Thus, T-cellmediated cytotoxicity against virus-infected (5) or hapten-modified cells (6) and against minor histocompatibility locus antigens (7) requires syngenicity between target and original stimulator cells with respect to K or D regions. On the other hand, proliferative responses to protein antigens are facilitated by I region homology between interacting macrophages and lymphocytes in both guinea pigs and mice (8, 9). Secondary proliferative responses to haptenmodified cells may also be facilitated by homology within H - 2 (10). Recently Newman et al. (11) have reported that primed human lymphoid cells respond to hapten-conjugated lymphoid cells in vitro. Using a similar system, the present report describes development of a human model to study both primary and secondary proliferative responses to trinitrophen.ylated human cells. Moreover, the data connote a role for the human MHC in T-cell recognitive processes. Human lymphocytes primed to hapten-modified autologous peripheral blood mononuclear cells gave substantially greater proliferative responses when restimulated with autologous than with allogeneic hapten-conjugated cells. HLA-A and HLA-B locus antigens do not appear to participate in such restimulation, whereas B-cell typing studies strongly suggest a role for gene products of the HLA-D region in definition of functional immunogenic units. * Supported by National Institutes of Health grants AI 13810 and RR 00350. Computational assistance was provided by the CLINFOProject, National Institutes of Health Contract NO1-RR5-2118. $ Investigator, HowardHughes MedicalInstitute. 1Abbreviations used in this paper: HBSS, Hanks' balanced salt solution; MHC, major histocompatibilitycomplex; MLC, mixed leukocyte reaction; PBMCs, peripheral blood mononuclear cells; TNP, trinitrophenyl. J. ExP. MED.© The RockefellerUniversity Press. 0022-1007/78/0601-167151.00




M a t e r i a l s a n d Methods Peripheral Blood Mononuclear Cells (PBMCs). Heparinized venous blood was obtained from healthy h u m a n subjects and PBMCs were isolated by flotation on sodium diatrizoate-Fieoll cushions (Isolymph; Teva, Ltd., Jerusalem, Israel) by previously described techniques (12). Cells were washed twice with Hanks' balanced salt solution (HBSS; Microbiological Associates, Walkersville, Md.) before hapten conjugation or cell culture. Cells obtained in this fashion contain approximately 80% lymphocytes and 20% monocytes. An average of 70% of these cells were classified as T cells on the basis of E-rosetting with sheep erythrocytes. Trinitrophenyl (TNP)-Conjugated Cells. Hapten-modified PBMCs were prepared by modifications of the technique described by Shearer for murine cells (13). Briefly, 5-25 × 10e cells/ml were incubated 12 min at 37°C with 2 mM trinitrobenzene sulfonic acid (Nutritional Biochemicals Corp., Cleveland, Ohio) in HBSS buffered with 50 mM Hepes (Microbiological Associates) and titrated to pH 7.8 with 1 N NaOH. Cells were washed three times with Hepes-buffered HBSS supplemented with 10 mg/ml glycylglycine (Grand Island Biological Co., Grand Island, N.Y.) and 5% heat-inactivated fetal calf serum (Grand Island Biological Co.) to remove all unreacted trinitrobenzene sulfonic acid. Cell Culture Techniques. Cells were cultured in RPMI-1640 (Grand Island Biological Co.) supplemented with 20% AB positive heat-inactivated pooled h u m a n plasma, 2 mM L-glutamine (Microbiological Associates) and 50 #g/ml gentamicin (Schering Corp., Kenilworth, N.J.). Cultures were buffered to pH 7.2 with 25 mM Hepes and 1 N NaOH and were incubated in a 5% CO2 atmosphere at 37°C. For primary stimulation cells were cultured in either round-bottom microculture wells (1SMRC-96; Linbro Scientific Inc., Becton, Dickinson & Co., New Haven, Conn.) or in round-bottom 17 × 100-mm tubes (2001; BioQuest, BBL, & Falcon Products, Cockeysville, Md.). In general, tubes were utilized in long-term cultures for ease in media replacement. Responder and stimulator cells were each cultured at a final concentration of 7.5 × 105 cells/ml in cultures of 0,2 ml (microculture wells) or 2 ml (tubes). For primary sensitization the cells were cocultured for 21-30 days. Primary proliferative responses were monitored from 3 to 25 days after culture initiation. Stimulator cells were treated with 500 or 4,000 rads 137Cs (dose rate 6,000 rads/min) or with mitomycin C (Sigma Chemical Co., St. Louis, Mo., 50/zg/ml for 30 min at 37~C). Inactivated stimulator cells are designated throughout with a subscript x. Culture media were replaced with fresh media once each week. For secondary restimulation, primed cells were harvested, washed once with medium, and cocultured in microculture wells at a final concentration of 7.5 x 105 viable cells/ml, with an equal concentration of fresh stimulator cells. Stimulator cells in secondary cultures were either TNP-conjugated or unconjugated, and in all experiments were inactivated with 4,000 rads Tirradiation. Proliferative responses were assessed by addition of 1.0 /zCi of tritiated thymidine (sp act 2.0 Ci/mM; New England Nuclear Corp., Boston, Mass.) to cultures for the final 20 h of the incubation period. HLA Typing. Typing for HLA-A and HLA-B locus antigens was by standard serologic techniques, employing the microcytotoxicity assay (14). All serologic procedures were performed by personnel of the Methodist Hospital tissue typing laboratory. B-Cell Antigen Typing. B-cell antigens were detected by using the method o f T i n g et al. (15). Briefly, a B-cell-enriched population was prepared by rosetting PBMCs with neuraminidasetreated sheep erythrocytes over sodium diatrizoate-Ficoll cushions. The non-erythrocyte rosetting (interface fraction a ~ e r centrifugation) B-cell-enriched population was then typed for B-cell antigens by using anti-B cell antisera trays kindly provided by Dr. Paul Terasaki, University of California at Los Angeles, utilizing techniques similar to A and B locus typing. Data Analysis. Data from separate experiments are expressed as mean cpm of three to four replicate cultures with the SEM. Net counts per minute (E-C) were calculated by subtracting cpm of responses to unconjugated stimulators (C) from cpm of cultures with TNP-conjugated stimulator cells (E). E-C errors were determined by the formula for propagation of error. Stimulation indices (E/C) were calculated by dividing (E) by (C) values as defined above. Two-tailed rank-sum tests were performed to determine significant differences between experimental groups.





Primary Responses of Fresh PBMCs to Autologous TNP-Conjugated Cells* High responders$ I



798 2,116 377 587

± 139 ± 984 ± 48 ± 139

Low responders G N L K

1,310 566 493 322

-+ -+ ± -+


198 151 49 79

TNP-Conjugated 20,813 17,920 7,533 7,306

4,520 2,213 871 997

± 1,781 ± 1,768 z 822 ± 726

± 644 ± 682 ± 319 -+ 50

E-C 20,015 15,804 7,156 6,718

3,210 1,646 378 675

± 1,951 ± 2,023 ± 825 ± 735

x 674 ± 698 _ 323 _+ 105

E-C Rangell 11,370 6,746 6,760 5,842

- 20,015 - 19,079 - 15,507 - 6,718

194 1,104 187 -472

- 4,087 - 1,646 - 378 - 675

* 1.5 x 105 r e s p o n d e r cells, cocultured w i t h 1.5 × 105 u n c o n j u g a t e d or T N P - c o n j u g a t e d i n a c t i v a t e d a u t o l o g o u s cells, were pulsed w i t h [:~H]thymidine for 20 h 5-6 d a y s a f t e r c u l t u r e in m i c r o t i t e r wells. $ H i g h a n d low r e s p o n d e r s were d i s t i n g u i s h e d u s i n g 2-7 s e p a r a t e e x p e r i m e n t s for e a c h i n d i v i d u a l . E a c h low r e s p o n d e r w a s a s s a y e d w i t h a h i g h r e s p o n d e r in at l e a s t two e x p e r i m e n t s . H i g h r e s p o n d e r s in all e x p e r i m e n t s h a d E/C v a l u e s > 5, w h e r e a s low r e s p o n d e r s in all e x p e r i m e n t s h a d E/C v a l u e s < 5. § D a t a are t a k e n from r e p r e s e n t a t i v e e x p e r i m e n t s w i t h triplicate or q u a d r u p l i c a t e d e t e r m i n a tions. II E-C r a n g e of all e x p e r i m e n t s p e r f o r m e d for e a c h subject.

Results Primary TNP-Conjugate Responses. In vitro coculture of PBMCs with autologous TNP-conjugated cells stimulated primary proliferative responses in 16 of 18 individuals tested. Patterns of relative responsiveness were consistent and clearly differentiated high and tow responders (Table I). 5 of 18 subjects tested were classified as low responders in that they always had stimulation indices of less than 5. High responders had an average stimulation index of 11 _+ 1 (E/C _+ SEM) compared with 2.1 _+ 0.3 for low responders. This difference was not due to a shift in response kinetics, nor was it due to a general decrease in response capacity of low responder subjects in that high and low responders exhibited comparable proliferative responses to phytohemagglutinin and delayed skin test reagents (D. L. Peavy and M. F. Seldin, unpublished observations). Studies of the effect of stimulator cell inactivation on primary proliferative responses to TNP-conjugated cells revealed no difference in the proliferative responses of lymphocytes from high responder subjects to stimulators inactivated with 4,000 rads compared with 500 rads. In contrast, low responder lymphocytes, did not respond to high-dose irradiated stimulator cells, although marginal but significant responses were generated with cells from three out of five low responder subjects when stimulator cells were not inactivated or when they were inactivated with 500 rads or with mitomycin C. Secondary Responses. Restimulation of primed responder cells 3 wk after initial sensitization generated a secondary proliferative response to TNP-conjugated autologous PBMCs. The kinetics of the primary and secondary responses were easily distinguishable (Fig. 1). Primary responses showed little if any







m \\.'~-. ~In \\~. "~I II\ \

FC ; \ \ ~ 60





6 (DAYS)




FIG, 1. Kinetics of primary and secondary proliferative responses to hapten-modified autologous cells. Data were taken from three separate experiments and normalized to percent maximal E-C response.

proliferation at day 3 and peaked at day 8; in contrast, secondary responses were maximal at day 3 and had diminished substantially by day 6. The peak secondary response was usually greater than the maximum primary response seen at a later time point. The stimulation index of secondary responses varied from 6.5-20 and averaged 14-fold at day 3. PBMCs precultured for 3 wk without stimulator cells generated responses consistent with primary responses, both in magnitude and kinetics, when TNP-conjugated stimulator cells were added after the preincubation period. In general, cells from subjects which gave low primary responses failed to give responses after secondary restimulation. Low primary responder G (Table I), however, gave both primary and secondary responses when primary stimulators were mitomycin C-treated (Table II) or not inactivated (Table III), although no secondary response was seen when primary stimulators were treated with 4,000 rads. Secondary responses were not affected by inactivation of secondary stimulators with the high dose (4,000 rads) ofT-irradiation. Preferential Restimulation with Autologous TNP-Conjugated Cells. To investigate whether secondary TNP-conjugate responses were dependent on recognition of some autologous determinants in addition to the hapten moiety, primed respenders were restimulated with both autologous and heterologous TNP-modified PBMCs. The results of one such experiment are seen in Table II. In this experiment a self-preference in the secondary response was seen. That is, the response to autologous TNP-conjugated stimulators was much greater than that to allogeneic TNP-conjugated stimulators. In contrast to data reported in the mouse, in which cells primed with TNP-conjugated autologous cells were unresponsive to alloantigens (10), the response of autologous TNP-conjugate primed cells to alloantigen was often accelerated. Furthermore unprimed precultured cells developed proliferative responses following the usual kinetics of human MLC responses. Influence of HLA-A and HLA-B Locus Determinants on Secondary TNPConjugate Responses. We next considered whether MHC determinants were involved in the recognition of TNP-conjugated cells. Fig. 2 shows a single experiment in which two responders were primed to autologous TNP-conjugated PBMCs and then restimulated with the same panel of three stimulators who







R. R I C H


Self-Preference of Secondary Response to Hapten-Modified PBMCs Stimulators* (HLA Phenotype) A B G (1,2

27,40) H



G Responders$ Unprimed Unconjugated TNP-Conjugated Unconjugated TNP-Conjugated

227 401 1,945 1,041


_+ 34 -+ 71 -+ 146 -+ 164

756 10,998 5,841 5,877

_+ 69 + 719 _+ 146 _+ 403

* TNP-conjugated or unconjugated stimulators were fresh P B M C s inactivated with 4,000 rads and cultured at 1.5 × 105/microtiterwell. Cultures were harvested 68 h after secondary stimulation. $ Responders were cultured at 1.5 × 105 cells/microtiterwell. Unprimed responders were fresh PBMCs. Primed responders were cocultured 3 wk with mitomycin C-inactivated trinitrophenylated autologous cells. 12






8 0 X


CL (J 4

STiMULAtOrS ,~ HLA ~ B PHENOTYmES ' 29, W~'~-~,?


J e (2~r71




• s (29,W~,7)





.LA .o~o~.oG, W,'H ~ESPONOEP

SELF - -+ - SELF ++ FIG. 2, H L A associationof secondary responses. Primed responders were cocultured with 500 rad inactivated TNP-conjugated autologous cells and restimulated with 4,000 rad inactivated TNP-conjugated or unconjugated cells. Mean cpm -+ S E M of responses to conjugated cells are represented by the total height of each bar, with the response to unconjugated cellsshown in the cleararea. As will be subsequently discussed (seetext and Table IV) stimulator O shares a D region antigen (DRw2) with Responder A, in addition to the HLA-B7 antigen. had been typed for H L A - A and H L A - B locus serologic specificities. For both responders there w a s a proliferative response to TNP-conjugated allogeneic cells in excess of that to alloantigens alone. T h e level of proliferation of one responder to TNP-conjugated cells that shared both H L A - A antigens w a s the s a m e as the response to a completely HLA-nonidentical stimulator. That is, the responses of primed cells from subject D to stimulator cells O a n d A were essentially equivalent a n d were about 3 0 % of that to autologous TNP-conjugated cells (E-C analysis). T h e response of the other responder (A) s h o w e d a m u c h higher level of proliferation to allogeneic TNP-conjugated cells that shared an H L A - B antigen (O) than to completely HLA-nonidentical stimulator cells (D). A further examination of the functional role of H L A - A and H L A - B antigen homologies in secondary TNP-conjugate responses w a s therefore performed (Table III). T h e presence or absence of H L A - A antigen h o m o l o g y did not affect

9,181 _+ 1,220 ND


4,634 _+ 641




19,436 ± 2,645










HLA Homology

Primed M r e s p o n d e r $



750 ±



361 ± 1,183

15,588 ± 1,975




E-C ± SEM§

Primed G responder* HLA Homology A-l,2






3,873 ± 530 6,008 ± 372


213 ± 401

2,105 ± 453

6,074 ± 258

2,944 + 255









Primed O responder HLA Homology

* G primed responders were from c u l t u r e s with autologous TNP-conjugated cells t h a t were not inactivated. $ M primed and O primed responders were from cultures with autologous 4,000 rad inactivated T N P stimulators. § Secondary cultures of all responders were harvested 68 h after culture initiation.


GTNP, 27,40) BTNP~ (10,W32 5,W35) MTNPx (2,3 12,40) OTNP~ (2,25 7,18) JTNP~ (1,3 5,8) PTNP, (2,W31 5,12) TTNPx (W24,W31 7,14) CTNP~ (1,9 7,8)

TNP-Conjugated stimulators (HLA phenotype) A B


Effect of HLA A and B Locus Homology on Secondary Responses




TABLE IV DRw Typing of Subjects Subjects

B-Cell antigens*


No antigens detected DRwl (weak reaction) Not typed DRw2, DRwl, DRw6 DRw2 DRw3 DRw7 DRw7 DRw2, DRw3 (weak reaction) DRw2 DRw7, DRw3 (weak reaction)

* 1977 WHO nomenclature for D-region related (DR) antigen (17).

the magnitude of responses, whereas HLA-B antigen homology correlated in some but not all, combinations with higher responses. One stimulator (C) which shared an HLA-B antigen with the primed responder (O) gave a proliferative response equivalent to autologous conjugated cells. Influence of HLA-D Region Determinants on Secondary TNP-Conjugated Respones. In each experiment in which HLA-B antigen homology between stimulator and responder cells resulted in augmented responses the antigen shared was HLA-B7. Our attention was therefore directed to the HLA-D region, since HLA-B7 is in strong linkage disequilibrium with Dw2 (16). Two alternative methods, MLC responses to homozygous typing cells (Dw specificities) and serologic detection of B-cell antigens can be used to determine D region alloantigens. The B-cell antigens utilized in our investigation have been designated DRw antigens, corresponding to the Dw specificities detected by the functional tests (WHO Conference 1977 [17]). In the two examples where HLA-B antigen homology was clearly associated with a higher response than other allogeneic stimulators (Fig. 2, Responder A with Stimulator O: Table III, Responder O with Stimulator C) the individuals also shared the DRw2 antigen (Table IV). In addition the DRw2 antigen was also shared (Table III, Responder O) with a stimulator (M) that was not as effective as a DRw2 negative subject (T) in restimulating the response. It may be relevant in this regard that subject M was Black, whereas with the other DRw homologies the subjects were Caucasian. More data relevant to the question of HLA-D region antigen association of the secondary proliferative response are shown in Table V. Stimulators that shared HLA-D region antigens with the responder cells generally gave higher responses, expressed as a percentage of the response with autologous stimulators (76.5 _+ 5.3%; mean _+ SEM), than did those that did not share these antigens (42.0 _ 7.0%). It should be noted that in experiment 1, stimulator Z was the son of responder A. Thus haploidentity in'the one experiment where this was examined was not more effective than HLA-D antigen homology between allogeneic individuals. The data presented in this table also illustrate the finding that the response generated by allogeneic TNP-conjugated cells was




12, 17)


















HLA Homology






31 _+ 1

39-+ 3

60 -+ 1

69 +- 3

88 -+ 2

Exp. 1§ 100 -+ 1

17 -+ 3

64 +_ 1

Exp. 2 100 +_ 8










+ Self





Antigen homology

HLA Homology A B

HLA-D % Maximal~ E-C + SEM -

Antigen* homology

44 +_ 1

57-+ 4

53 -+ 2

71 +- 3

100 ± 1

Exp. I 83 +-- 2

% Maximal E-C +- SEM

P r i m e d 0 Responders


Primed A Responders


99 +- 6

100 _+ 4

Exp. 3Jl

(cpm allogeneic unconjugated) x 100%. (cpm syngeneic unconjugated)]

responses were harvested at 68 h. ¶ S t i m u l a t o r Z is the son of Responder A.

Secondary cultures were h a r v e s t e d at 48 h. [I In e x p e r i m e n t 3 primed O r e s p o n d e r s were generated by coculture with 4,000 rad inactivated autologous t r i n i t r o p h e n y l a t e d cells. Secondary

§ In e x p e r i m e n t s 1 + 2 primed r e s p o n d e r s (A and O) were g e n e r a t e d by coculture w i t h 500 rad inactivated autologous t r i n i t r o p h e n y l a t e d cells.

(cpm allogeneic TNP-conjugated) [(cpm syngeneic TNP-conjugated)

Except for s t i m u l a t o r Z (see footnote ¶) the HLA-D a n t i g e n homology between r e s p o n d e r and s t i m u l a t o r w h e r e indicated w a s in each case the DRw2 antigen (see Table IV), reflecting the s t r o n g linkage d i s e q u i l i b r i u m between HLA-B7 and this D region allele. Percent m a x i m a l E-C w a s calculated by d e t e r m i n i n g the m e a n r e s p o n s e to s t i m u l a t o r s from triplicate or q u a d r u p l i c a t e c u l t u r e s and employing the following formula:

(1,W30 TTNP~ (W24,W3 DTNP~ (2,25



ATNPx (29,W30 5,7) OTNP x


Stimulator HLA phenotype


HLA-D Antigen Association of Secondary Proliferative Responses


O ~







M I C H A E L F. S E L D I N A N D R O B E R T R . R I C H


not entirely explained on the basis of shared MHC specificities, insofar as they were detected. HLA-D region determinant homology between primed responder and stimulator correlated significantly with the magnitude of the secondary TNP-conjugate proliferative response. The hypothesis that the two samples (those sharing and not sharing HLA-D antigens) were drawn from the same population was rejected (P < 0.01; two-tailed rank-sum test). On the other hand, stimulators with HLA-A or HLA-B homology with the responder (excluding stimulators where HLA-D antigen homology with the responder was also present) were not significantly different from stimulators with no responder homology (P > 0.2). Discussion This investigation was undertaken to develop an in vitro model for examination of the functional significance of MHC determinants in human immune responses. We studied proliferative responses of human lymphoid cells which had been conjugated to a well-defined hapten (TNP), based on the previous findings of MHC association in murine responses (10). Recently methods for obtaining primed proliferative and cytotoxic responses to hapten-modified human lymphoid cells have been reported (11, 18). Although self-preference was suggested, the role of the MHC, if any, in these responses was not established. Our experiments demonstrate that under appropriate conditions human lymphocytes develop both primary and secondary responses in vitro to TNPconjugated lymphoid cells. Primary proliferative responses have not been previously reported for either human or murine TNP-conjugate systems. Based on primary proliferative responses to TNP-conjugated autologous cells, individuals could be segregated into high and low responders. Five of 18 subjects studied were classified as low responders. Three of these low responders generated minimal primary responses but were extremely sensitive to stimulator cell inactivation. A likely explanation for this observation is that functional stimulator cells are required in greater numbers or for a longer time period to trigger responses in these individuals. This in turn may be caused by decreased efficiency in antigen presentation, or recognition, or beth. Although we can only speculate on mechanisms responsible for high and low responsiveness, it is unlikely that previous exposure (priming) to TNP had occurred. Furthermore these same subjects did not segregate into high and low responders when stimulated with mitogens or delayed skin test reagents. Similar phenomena in proliferative responses to soluble antigens (4) and cytotoxic responses to TNP-conjugated cells (6) in the mouse have been explained on the basis of Ir genes encoded within the I region of H-2, which is analogous to the HLA-D region. The possibility that high and low responsiveness to TNP-conjugates in humans may reflect presentation or recognition of particular D region gene products, rather than responsiveness to TNP per se, is therefore particularly interesting. Responsiveness to other haptens in the same system may resolve this possibility. A major interest has been the possible involvement of MHC gene products in stimulation of secondary responses. The data clearly demonstrate self-preference in the secondary stimulation phase, although various levels of responsive-




ness to allogeneic TNP-conjugated ceils were seen. Several possible mechanisms were considered to explain both self-preference and apparent reactivity to TNP on allogeneic ceils in the secondary response. (a) Since alloantigens can stimulate suppressor T-cell function (19, 20) the apparent self-preference of the secondary response may be due to the absence of alloantigen determinants rather than a requirement for autologous structures. We consider this explanation unlikely because of the accelerated kinetics of the secondary response. Furthermore, preliminary experiments demonstrate that the addition of allogeneic cells to secondary cultures does not suppress responses to autologous TNP-conjugated stimulators. (b) Part or all of the response may be to TNPprotein on the cell surface without special regard to HLA. Self-preference may, therefore, be due to autologous proteins having unique determinants. While not formally excluded, this possibility conflicts with the findings in murine models and would not explain our findings of preferential restimulation with partial HLA-D region homologous stimulators. (c) There may be a requirement comparable to that in murine systems for certain MHC determinant homologies. Our data did strongly suggest that D region homology between primed responder and stimulator account for increased responsiveness. It should be stated, however, that the limited number of D region alleles which can be currently detected constrain this conclusion. It is possible, of course, that the requisite genetic homologies between responder and stimulator cells are not with detected alloantigenic determinants, but with other HLA-D region gene products in linkage disequilibrium with the typed alleles. In murine secondary proliferative responses to TNP-conjugated cells a strict MHC homology restriction was seen, but intra-H-2 mapping was complicated (10). K + I region homology was more effective in restimulating secondary responses than I region homology alone. Since all subjects in this study with DRw2 homology also shared the HLA-B7 antigen a similar cooperative interaction between HLA-B and D region determinants must be considered. It must be emphasized that allogeneic TNP-conjugated cells which did not share any detected MHC determinants with the primed responder still frequently stimulated a measurable proliferative response, albeit substantially less than autologous cells. The role of the human MHC in T-cell-mediated responses to conventional antigens is not yet clear. Differences between apparent mouse and human homology requirements may be explained by more extensive crossreactivities between human than routine MHC determinants (i.e., differences between outbred and inbred populations). Recent work with wild mice has shown that cross-reactivities in the TNP cell-mediated lympholysis assay do occur in 1"1-2 disparate hosts when the tl-2 haplotypes between stimulator and target cells are closely related (21). A less likely explanation is the possibility that the recognitive structures in human lymphocytes are less discriminating than murine receptors. That is, greater cross-reactivity at the receptor level rather than the stimulator determinant level could account for disparities. Alternatively, the apparently nonrestricted component of the response may represent proliferation induced by TNP-conjugated allogeneic proteins presented by syngeneic macrophages. Since syngeneic macrophages are present in









both the primary and secondary phase, MHC restriction requirements would not preclude a proliferative response dependent on macrophage handling of TNP-conjugated proteins on the cell surface. The much lower percentage of macrophage-monocytes in murine spleen cell populations (3-5%) than in human PBMC populations (20-25%) could explain the strict homology requirements in secondary TNP-conjugate proliferative responses reported in the mouse (10). Recently Dickmeiss et al. (22) reported that cell-mediated cytotoxicity against hapten-conjugated lymphocytes is HLA restricted. They utilized dinitrophenylated PBMCs from subjects who had been contact-sensitized in vivo with dinitrochlorobenzene or dinitrofluorobenzene. Aider an additional in vitro sensitization phase, the cytotoxic responses were largely restricted to H L A - A or H L A - B locus homologous hapten-conjugated targets. Thus, their findings with cytotoxicity contrast with our results that human proliferative responses to hapten-conjugated cells are associated with the HLA-D region, and both sets of data are in accord with previous studies in mice (6, 10). The role of HLA determinants in recognition and/or presentation of conventional antigens may help elucidate the pathogenesis of a variety of diseases. The association of a number of human diseases with certain H L A alleles, probably a reflection of linkage disequilibrium, has become apparent in recent years (1, 2). Moreover, juvenile diabetes mellitus (23, 24) and ragweed hayfever (25), which are not associated with particular H L A alleles but segregate with the MHC in family studies, may be caused in part by expression of Ir genes encoded in the D region. We found that secondary proliferative responses of human lymphocytes to TNP-conjugated lymphoid cells are associated with the HLA-D region. Thus the model we describe has important implications for definition of human T cell recognitive units and may permit a functional characterization of the role of HLA determinants in both normal and pathologic states. Summary An in vitro model was developed to study both primary and secondary proliferative responses of human lymphocytes to hapten-conjugated peripheral blood mononuclear cells. Coculture of human lymphocytes with autologous trinitrophenyl (TNP)-conjugated stimulator cells resulted in primary proliferative responses. Subjects segregated into high and low primary responders with mean stimulation indices of 11 and 2.1, respectively. Restimulation of primed cells from high responder subjects 3 wk after initial sensitization generated secondary proliferative responses. To investigate the antigenic requirements for secondary stimulation, autologous TNP-conjugate primed respenders were restimulated with both autologous and allogeneic TNP-conjugated stimulators. In all experiments restimulation with autologous conjugated cells yielded substantially greater proliferative responses than with allogeneic conjugates. Experiments were then performed to ascertain whether HLA determinant homology between primed responder and stimulator cells influenced the level of secondary responsiveness. Homology for HLA-A and B locus serologic determinants was not associated with enhanced responsiveness. In contrast, D region determinant homology, detected by B-cell



a n t i g e n typing, showed a highly significant positive correlation with the m a g n i t u d e of secondary responses. The data t h u s strongly suggest t h a t for secondary proliferative responses to TNP, h u m a n T cells recognize h a p t e n in association with HLA-D region determinants. We wish to thank Shirley Qualia and Kathy Freeman for excellent technical assistance in HLA serologic and B-cell antigen typing and Dr. Paul I. Terasaki and Mr. Ken Slyker for making Bcell typing trays available. The advice of Dr. Wes Bullock and Dr. Joe Templeton and the assistance of Dan Kastner in data analysis are gratefully acknowledged. We are indebted to Mrs. Elaine Bowers for her perseverance in the preparation of this manuscript.

Received for publication 17 January 1978. References 1. Svejgaard, A., P. Platz, L. P. Ryder, L. S. Nelson, and M. Thomsen. 1975. HL-A and disease association-a survey. Transplant. Rev. 22:3. 2. Bach, F. H., and J. J. van Rood. 1976. The major histocompatibility complexgenetics and biology. N. Engl. J. Med. Part III. 295:927. 3. McDevitt, H. O., M. B. A. Oldstone, and T. Pincus. 1974. Histocompatibility-linked genetic control of specific immune responses to viral infection. Transplant. Rev. 19:209. 4. McDevitt, H. O., and B. Benacerraf. 1969. Genetic control of specific immune responses. Adv. Immunol. 11:31. 5. Doherty, P. C., R. V. Blanden, and R. M. Zinkernagel. 1976. Specificity of virusimmune effector T cells for H-2K or H-2D compatible interactions: Implications for H-antigen diversity. Transplant. Rev. 29:89. 6. Shearer, G~ M., T. G. Rehn, and A.-M. Schmitt-Verhulst. 1976. Role of the murine major histocompatibility complex in the specificity of in vitro T cell-mediated lympholysis against chemically-modified autologous lymphocytes. Transplant. Rev. 29:222. 7. Bevan, M. J. 1975. The major histocompatibility complex determines susceptibility to cytotoxic T cells directed against minor histocompatibility antigens. J. Exp. Med. 142:1349. 8, Yano, A., R. H. Schwartz, and W. E. Paul. 1977. Antigen presentation in the murine T-lymphocyte proliferative response. I. Requirement for genetic identity at the major histocompatibility complex. J. Exp. Med. 146:828. 9. Rosenthal, A. S., and E. M. Shevach. 1976. Macrophage-T lymphocyte interaction: The cellular basis for genetic control of antigen recognition. In The Role of Products of the Histocompatibility Gene Complex in Immune Responses. D. H. Katz and B. Benacerraf, editors. Academic Press, Inc. New York. p. 335. 10. Schmitt-Verhulst, A.-M., G. A. Garbino, and G. M. Shearer. 1977. H-2 homology requirements for secondary cell-mediated lympholysis and mixed lymphocyte reactions to TNP-modified syngeneic lymphocytes. J. Immunol. 118:1420. 11. Newman, W., G. L. Stoner, and B. R. Bloom. 1977. Primary in vitro sensitization of human T cells. Nature (Lond.). 269:151. 12. Boyum, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest. 21(Suppl. 97):77. 13. Shearer, G. M. 1974. Cell-mediated cytotoxicity to trinitrophenyl-modified syngeneic lymphocytes. Eur. J. Immunol. 4:527. 14. Staff, Research Resources Branch, NIAID. 1976. NIH lymphocyte microcytotoxicity technique. NIAID Manual of Tissue Typing Techniques. ~U.S. Department of Health, Education and Welfare). NIH. 78-545:22.



15. Ting, A., M. M. Ray, and P. I. Terasaki. 1976. B-lymphocyte alloantigens in Caucasians. J. Exp. Med. 143:981. 16. Albert, E. D., W. Mempel, and H. Grosse-Wilde. 1973. Linkage disequilibrium between HL-A7 and the MLC specificity Pi. Transplant. Proc. 5:1551. 17. Bodmer, W. 1977. Histocompatibility Testing. Munksgaard, Copenhagen. In press. 18. Levis, W. R., and A. M. Dattner. 1977. Clonal priming oflymphocytes can measure human immune response gene function. N. Engl. J. Med. 297:842. 19. Rich, S. S., and R. R. Rich. 1974. Regulatory mechanisms in cell-mediated immune responses. I. Regulation of mixed lymphocyte reactions by alloantigen-activated thymus-derived lymphocytes. J. Exp. Med. 140:1588. 20. Sondel, P. M., M. W. Jacobson, and F. H. Bach. 1975. Pre-emption of human cellmediated lympholysis by a suppressive mechanism activated in mixed lymphocyte cultures, J. Exp. Med. 142:1606. 21. Forman, J., and J. Klein. 1977. Immunogenetic analysis of H-2 mutations. VI. Crossreactivity in cell-mediated lympholysis between TNP-modified cells from H-2 mutant strains. Immunogenetics. 4:183. 22. Dickmeiss, E., B. Soeberg, and A. Svejgaard. 1977. Human cell-mediated cytotoxicity against modified targets is restricted by HLA, Nature (Lond.) 270:526. 23. Rubinstein, P., N. Suciu-Foca, and J. F. Nicholson. 1977. Close genetic linkage of HLA and juvenile diabetes mellitus. N. Engl. J. Med. 297:1036. 24. Barbos, J., R. King, H. Noreen, and E. J. Yunis. 1977. The histocompatibility system in juvenile insulin-dependent diabetic multiplex kindreds. J. Clin. Invest. 60:989. 25. Blumenthal, M. N., D. B. Amos, H. Noreen, N. R. Mendell, and E. J. Yunis. 1974. Genetic mapping of Ir locus in man: linkage to second locus of HLA. Science (Wash. D.C.). 184:1301.

Human immune responses to hapten-conjugated cells. I. Primary and secondary proliferative responses in vitro.

H U M A N I M M U N E R E S P O N S E S TO H A P T E N - C O N J U G A T E D CELLS I. P r i m a r y a n d S e c o n d a r y P r o l i f e r a t i v e...
822KB Sizes 0 Downloads 0 Views